Deep Learning with PyTorch

About this course

A from-first-principles course in deep learning using PyTorch. You'll write and train networks by hand before reaching for pretrained models, so the architectures in the second half (CNNs, attention, transfer learning) actually make sense.

What you'll learn

• Build and train neural networks in PyTorch from scratch
• Understand backpropagation well enough to debug it
• Apply CNNs and attention-based models to real tasks
• Use transfer learning to get results with less data
• Train on GPUs and deploy a trained model

Student reviews